Geometric Interpretation and Optimization of Large Semantic Data Sets in Real-Time VR Applications

Roßmann

Sondermann

2013

ELCVIA

Self Cite

During the development process of new algorithms for computer vision applications, testing and evaluation in real outdoor environments is time-consuming and often difficult to realize. Thus, the use of artificial testing environments is a flexible and cost-efficient alternative. As a result, the development of new techniques for simulating natural, dynamic environments is essential for real-time virtual reality applications, which are commonly known as Virtual Testbeds. Since the first basic usage of Virtual Testbeds several years ago, the image quality of virtual environments has almost reached a level close to photorealism even in real-time due to new rendering approaches and increasing processing power of current graphics hardware. Because of that, Virtual Testbeds can recently be applied in application areas like computer vision, that strongly rely on realistic scene representations. The realistic rendering of natural outdoor scenes has become increasingly important in many application areas, but computer simulated scenes often differ considerably from real-world environments, especially regarding interactive ground vegetation.In this article, we introduce a novel ground vegetation rendering approach, that is capable of generating large scenes with realistic appearance and excellent performance. Our approach features wind animation, as well as object-to-grass interaction and delivers realistically appearing grass and shrubs at all distances and from all viewing angles. This greatly improves immersion, as well as acceptance, especially in virtual training applications. Nevertheless, the rendered results also fulfill important requirements for the computer vision aspect, like plausible geometry representation of the vegetation, as well as its consistence during the entire simulation. Feature detection and matching algorithms are applied to our approach in localization scenarios of mobile robots in natural outdoor environments. We will show how the quality of computer vision algorithms is influenced by highly detailed, dynamic environments, like observed in unstructured, real-world outdoor scenes with wind and object-to-vegetation interaction.

Section: Distributing Ground Vegetationmentioning

confidence: 99%

Generation and Rendering of Interactive Ground Vegetation for Real-Time Testing and Validation of Computer Vision Algorithms

Roßmann

Sondermann

2013

ELCVIA

Self Cite

“…As shown on the right side of Figure 5, advanced lighting effects and a dynamic weather system including a rain simulation and a dynamic sky have been implemented, which match the current time and weather conditions. A new vegetation rendering approach was also developed, which creates high-quality real-time landscapes out of semantic datasets [9,10]. These close-to-reality virtual environments build an excellent basis for the development and testing of a broad range of eRobotic applications.…”

Section: B Integration and Interpretation Of Semantic Datasetsmentioning

confidence: 99%

“…These nodes are the basis for almost any render framework, containing elements necessary for basic rendering purposes. The active database approach is beneficial in order to optimize the render performance by identifying moving nodes, separate static and dynamic geometry and group together nodes with similar material and texture nodes [9]; however, the data usually received by GIS servers only contains basic geometry and textures , due to the fact that it is not built for rendering purposes. Directly rendered results mostly suffer from unpleasing visual quality as shown on the left side of Figure 5; however, all semantic data obtained has additional information describing the environment in more detail, which can be used to further increase the virtual environment's appearance.…”

Section: B Integration and Interpretation Of Semantic Datasetsmentioning

confidence: 99%

Taking the step from edutainment to eRobotics - A novel approach for an active render-framework to face the challenges of modern, multi-domain VR simulation systems

2013 IEEE Conference on E-Learning, E-Management and E-Services

Roßmann

2013

Self Cite

When talking about Virtual Reality (VR) today, most people just think of entertainment or education. Indeed, "edutainment", which is a mixture of the words entertainment and education, has always been a major domain of VR applications. VR simulation technology is also a well-known field of virtual training and engineering today and widely applied in research and development in a broad range of different domains. Due to the wide availability of realistic computer graphics in our day-to-day life, our eyes get spoiled by stunning computer generated virtual worlds nowadays; thus, the acceptance of these applications depends more and more on the realistic, state-of-the-art rendering of the virtual environment, as well as the convenient usage. While edutainment applications and models are specifically built to fulfill these requirements, the data, models and system structures of current VR simulation system in scientific domains usually are not. They are pure expert systems that demand for specific expertise in order to use them or understand the simulation results. The novel field of eRobotics can be understood as an eService related to advanced robotic applications and engineering. We will introduce novel system structures and techniques, which enable us to make the step from pure edutainment applications to complex but comprehensive eRobotics applications that can serve as a holistic development support tool in a broad range of different domains.

“…The results have been partly published in conference proceedings [129] [135] [128] [127] [276] [126] [279] [134] [125] [278] [272] [269] [273] [220] [274] [280] [268] as well as in journal articles [131] [130] [132] [133]. The results have been partly published in conference proceedings [129] [135] [128] [127] [276] [126] [279] [134] [125] [278] [272] [269] [273] [220] [274] [280] [268] as well as in journal articles [131] [130] [132] [133].…”

Section: Contributionmentioning

confidence: 99%

“…The VEROSIM 2 simulation system was an appropriate choice. All contained data is described through MetaTypes using a metamodel-based approach in order to give them an interpretable meaning [144] [134]. VEROSIM provides the software architectures and features required to implement the proposed concepts and approaches, which offers new possibilities to lift the system to a higher level, thus, making it capable of acting as a sustainable basis for a wide range of current and future eRobotics applications in multiple domains.…”

Section: Integration Into An Existing Simulation Systemmentioning

confidence: 99%

Bridging the Gap between Rendering and Simulation Frameworks

2016

Self Cite

the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.Printed on acid-free paper This Springer Vieweg imprint is published by Springer Nature The registered company is Springer Fachmedien Wiesbaden GmbH Dissertation RWTH Aachen, 2015 D 82 ForewordAfter nearly half a century of research, the Holy Grail of the field of computer graphics remains a comprehensive computational model and the corresponding software structure to provide realistically looking -and behaving -virtual world. However, Nico Hempe's work, documented in this thesis achieves nothing less than a functional and proven software structure to cope with the request for realism in the ever increasing complexity of today's robotic and mechatronic applications. Over the decades, modeling for computer simulation and visualization has progressed from an early reliance on purely geometric models to extremely powerful simulation models which also incorporate the underlying physical principles. Nico's is to date one of the most important PhD dissertations which manage to bridge the gap between advanced simulation and comprehensive rendering techniques by combining the best of both worlds to impressive virtual reality applications. His virtual worlds, which do not only look good but serve as the basis for virtual testbeds, are key components in new development approaches in the context of eRobotics and "Industry 4.0". Starting in 2010, Nico's work helped to establish the research field of eRobotics in order to cope with the inherent complexity, facilitate the development and significantly cut down the cost of advanced robotics and mechatronics applications. The objective is to effectively use electronic media -hence the "e" at the beginning of the term -to achieve the best possible advancements in the development of the robots' respective fields of use. Current domains of eRobotics covered by Nico's work comprise industrial robotics, work machines, construction site automation, space robotics and even the fields of environmental and city modelling. This thesis reveals the technical details behind a spectacular accomplishment in com...